Proliferation of mitochondria and gene expression of carnitine palmitoyltransferase and fatty...

Original article

Proliferation of mitochondria and gene expression of carnitinepalmitoyltransferase and fatty acyl-CoA oxidase in rat skeletalmuscle, heart and liver by hypolipidemic fatty acids

Geir K. Totlanda*, Lise Madsenb, Beate Klementsena,b, Hege Vaagenesb, Harald Kryvia, LivarFrøylandb,c, Sofie Hexebergb, Rolf K. Bergeb

a Department of Zoology, University of Bergen, Allégt. 41, 5007 Bergen, Norway

b Department of Clinical Biochemistry, University of Bergen, Haukeland University Hospital, 5021 Bergen,Norway

c Institute of Nutrition, Directorate of Fisheries, P.O. Box 185 Sentrum, 5804 Bergen, Norway

Received 23 March 2000; accepted 5 July 2000

Morphological and biochemical effects were induced at the subcellular level in the skeletal muscle, heart andliver of male rats as a result of feeding with EPA, DHA, and 3-thia fatty acids. The 3-thia fatty acid,tetradecylthioacetic acid (TTA) and EPA induced mitochondrial growth in type I muscle fibers in both thediaphragm and soleus muscle, and the size distribution of mitochondrial areas followed a similar pattern.Only the 3-thia fatty acid induced mitochondrial growth in type II muscle fibers. The mean area occupied bythe mitochondria and the size distribution of mitochondrial areas in both fiber types were highly similar inDHA-treated and control animals. Only the 3-thia fatty acid increased the gene-expression of carnitinepalmitoyltransferase (CPT)-II in the diaphragm. In the heart, however, the gene expression decreased. Inhepatocytes an increase in the mean size of mitochondria was observed after EPA treatment, concomitantwith an increase in mitochondrial CPT-II gene expression. Administration of 2-methyl-substituted EPA(methyl-EPA) induced a higher rate of growth of mitochondria than EPA. At the peroxisomal level in thehepatocytes a 3-thia fatty acid, EPA, and DHA increased the areal fraction concomitant with the inductionof gene expression of peroxisomal fatty acyl-CoA oxidase (FAO). In the diaphragm, mRNA levels of FAOwere not affected by EPA or DHA treatment, whereas gene expression was significantly increased after 3-thiafatty acid treatment. In the heart, both 3-thia fatty acid, EPA and DHA tended to decrease the levels of FAOmRNA. The areal fraction of fat droplets in all three tissue types was significantly lower in the groups treatedwith 3-thia fatty acid. In the group treated with EPA a lower areal fraction of fat droplets was observed, whilethe DHA group was similar to the control. This indicates that EPA and DHA have different effects onmitochondrial biogenesis. © 2000 Éditions scientifiques et médicales Elsevier SAS

3-thia fatty acids / omega-3-fatty acids / skeletal muscle / heart / liver / gene expression / carnitinepalmitoyltransferase / fatty acyl-CoA oxidase

* Correspondence and reprints: [email protected]: CMC, carboxymethyl cellulose; CPT, carnitine palmitoyltransferase; DHA, docosahexaenoic acid; EPA, eicosapentaenoicacid; FAO, fatty acyl-CoA oxidase; IMPs, intermembrane particles; PPAR, peroxisome proliferator-activated receptor; TD, dithiadicar-boxylic acid; TEM, transmission electron microscopy; TTA, tetradecylthioacetic acid

Biology of the Cell 92 (2000) 317−329© 2000 Éditions scientifiques et médicales Elsevier SAS. All rights reservedS0248490000010777/FLA

Proliferation of mitochondria and gene expression Totland et al.

1. INTRODUCTION

The triglyceride-lowering effect of fish oil (n-3 fattyacids) has been widely studied in mammals (Harris,1989). Evidence has been provided that the oxidativemetabolism of fatty acids is modified when rats werefed a diet enriched in EPA and DHA (Surette et al.,1992). The fish oils employed in a number of studiescontained both EPA and DHA, while only a few stud-ies have focused on the single-agent effects of EPA andDHA. These studies demonstrated that treatment withEPA, but not with DHA, decreased serum triglyceridelevels, and that this effect could be attributed in part toincreased hepatic mitochondrial fatty acid oxidation(Willumsen et al., 1996; Frøyland et al., 1997a; Ikeda etal., 1998; Madsen et al., 1999c). In contrast, both EPAand DHA increased peroxisomal �-oxidation in ratliver (Willumsen et al., 1993; Frøyland et al., 1996).Wheras EPA treatment mainly induces mitochondrialproliferation, DHA exclusively induces proliferation ofperoxisomes (Frøyland et al., 1997a). These results arein accordance with results from rat hepatocytes (Grønnet al., 1992; Berge et al., 1999; Madsen et al., 1999c).

The changes in oxidation of long-chain fatty acidsafter fish oil treatment may therefore be due to anincrease in the activity of the enzyme system respon-sible for both mitochondrial and peroxisomal�-oxidation. It is generally accepted that fatty acyl-CoAoxidase (FAO) is the rate-limiting enzyme for peroxiso-mal �-oxidation. The fatty acid oxidation that takesplace in the peroxisomes is independent of carnitine,but degraded long-chain fatty acyl groups are madeavailable for mitochondrial �-oxidation which is con-trolled by the carnitine palmitoyltransferase (CPT) sys-tem (see review by Ramsay, 1999). This process isprobably regulated by CPT-I in both liver and muscle(McGarry and Brown, 1997). CPT-II is not regarded asa rate-controlling factor in fatty acid oxidation. How-ever, in rat liver we have recently observed that CPT-II,but not CPT-I, was upregulated after treatment with3-thia fatty acids (Madsen and Berge, 1999a; Madsen etal., 1999b).

The 3-thia fatty acids are hypolipidemic sulphur-substituted fatty acid analogues, which affect bothmitochondrial and peroxisomal fatty acid oxidationboth in short term (hours) (Asiedu et al., 1993, 1995;Berge and Hvattum, 1994; Skrede et al., 1997; Vaageneset al., 1998) as well as long-term feeding (days, weeksand months) (Berge and Hvattum, 1994; Frøyland etal., 1995; Hexeberg et al., 1995; Asiedu et al., 1996). It isnot known whether 3-thia fatty acids, EPA and DHAhave any influence on these organelles in extrahepatictissues.

In order to address these questions, we selecteddiaphragm and soleus muscles, heart and liver for thestudy of: i) morphological changes in mitochondria in

distinct muscle fiber types, myocytes and liver; and ii)the gene-expression of rate-limiting enzymes involvedin fatty acid �-oxidation in the mitochondria and per-oxisomes. We also describe the changes that take placein the peroxisomal membrane during changes as out-lined above, and have selected hepatic peroxisomesfrom control and 3-thia fatty acid-fed animals as amodel for this process.

2. MATERIALS AND METHODS

2.1. Chemicals

Ethyl esters of docosahexaenoic acid (DHA), eicosa-pentaenoic acid (EPA) and 2-methyl EPA were pro-duced by Norsk Hydro AS, Research Center(Porsgrunn, Norway). The 3-thia fatty acid, tetrade-cylthioacetic acid (TTA) was prepared as describedpreviously (Berge et al., 1989; Spydevold and Bremer,1989) and the dithiadicarboxylic acid (1,10-bis(carboxymethylthio)decane) (TD) was prepared ac-cording to Aarsland et al. (1990). All other chemicalswere of highest reagent quality and were purchasedeither from Sigma Chemical Co. (London, UK) orMerck GmbH (Darmstadt, Germany). The purity andfatty acid composition of the supplements given to therats were identical to those described by Hexeberg etal. (1995).

2.2. Animals and treatments

Male Wistar rats from 140 to 160 g in weight werepurchased from Møllegaard Breeding Laboratory (Ejby,Denmark). They were housed individually in metalwire cages and allowed to acclimatize for 1 weekbefore the experiment started. The room was main-tained under a 12-h light:dark cycle at a temperature of20 ± 3 °C. All rats had free access to water and R34-EWOS-ALAB grower rat maintenance chow (12.5 MJ/kg) (EWOS, Sweden). The rats were randomlyassigned to one of six groups, i.e., DHA, EPA, 2-methylEPA, TTA, TD or a non-supplemented group receivingCMC (vehicle). Each group consisted of six animals.Twenty percent w/v stock suspensions of DHA, EPA,2-methyl EPA and 3 % w/v of TTA and TD wereprepared in 0.5 % CMC. Alpha-tocopherol (vitamin E)(0.5 % w/v) was added to the stock suspensions tominimize auto-oxidation. The fatty acid suspensionswere stored at sub-zero temperatures when not in use.

In the first experiment (long-term experiment),groups of rats were given either EPA, DHA or TTA byoro-gastric intubation for 3 months. The daily doseswere 1 g per kg body weight for EPA and DHA, and150 mg per kg for TTA. In the second experiment(short-term experiment), groups of rats were given 1.5g per kg body weight methyl-EPA, EPA, DHA or 300

318 Biology of the Cell 92 (2000) 317–329


mg per kg body weight TTA or TD daily for 10 days. Inthe third experiment, groups of rats were fed a stan-dard pellet diet with or without 0.3 % TD for 5 daysand used for morphological studies of peroxisomesfrom the liver. At the end of the experiments,overnight-fasted rats were anesthetized with HypnormDormicumt (fluanisone-fentanylmidazolam, 0.2mL·100 g–1 body weight) (Hyponorm; Janssens Phar-maceutica, Berse, Belgium; Dormicum; Hoffman, LaRoche AG, Basel. Switzerland). Blood samples werecollected in EDTA vacutainers after cardiac puncture.Diaphragm, soleus muscle, heart and liver were thenrapidly removed. The protocol was approved by theNorwegian State Commission for Animal Experimen-tation.

2.3. Morphological methods

2.3.1. Light- and electron microscopyDiaphragm, soleus muscle, the whole heart and liver

were carefully dissected out and immediately placed inice-cold fixative consisting of 0.2 M phosphate buffer(pH 7.35) containing 1 % formaldehyde, 2.5 % glutaral-dehyde, 0.08 M sucrose and 1.25 mM CaCl2 and with avehicle osmolarity of about 300 mOsm. After 15 minthin slices were taken from the latero-ventral part ofthe diaphragm, the mid-belly portion of the soleusmuscle, the anterior side of the left ventricle and fromthe liver. The thin slices from each of these tissues werethen cut into small slices, and systematic randomsampling was employed according to Gundersen et al.(1988). Five slices from each type of tissue from fiveanimals in each treatment group were used for themorphometric analysis. These slices were fixed in thealdehyde solution for a further 3 h before post-fixationin 1 % OsO4 in a vehicle with a composition identicalto that used during the aldehyde fixation. After 1 h offixation in OsO4 the specimens were dehydrated in agraded series of ethanol solutions and embedded inEpon (Fluka AG, Switzerland). From the diaphragmand soleus muscle semithin sections were stained inp-phenylenediamine (PPDA) (Korneliussen, 1972)which permits differentiation of mitochondria and fatdroplets. The differentiation between the two mainfiber types (type I and type II fibers) was based on thedistribution and content of mitochondria and fat drop-lets in addition to the cross-sectional fiber area andmyofibrillar pattern. In order to measure the samemuscle fibers at EM level, the ultrathin sections weretaken adjacent to the semithin sections ensuring thatthe EM morphometry was performed on representa-tive muscle fibers from each type established by thelight microscopic examination. The ultrathin sectionswere stained in uranyl acetate and lead citrate andstudied in a JEOL JEM-100 CX transmission electronmicroscope.

Five EM micrographs were selected from each sec-tion by means of a random number generator andexamined at a final magnification of × 8200. The meanareas of mitochondria and lipid droplets were mea-sured on a Leitz 64 K morphometric analyzer and theareal fractions (%) occupied by mitochondria and lipiddroplets (%) were calculated according to current mor-phometric principles (see review by Gundersen et al.,1988).

2.3.2. Purification and freeze-etchingof peroxisomes

The rats used for the morphological studies of puri-fied peroxisomes were tranquilized with CO2 and thenrapidly killed by decapitation to avoid accumulation ofblood in the liver. The liver was quickly excised andabout 1 g of liver tissue was homogenized in 4 mL ofan ice-cold sucrose medium consisting of 0.25 M su-crose, 10 mM HEPES, 1 mM Na-EDTA (pH 7.4). Acrude peroxisomal fraction enriched in peroxisomeswas obtained by differential centrifugation as de-scribed by De Duve et al. (1955) and Garras et al.(1995). One mL of the crude peroxisomal fraction wascarefully overlaid with 15 mL Nycodenz (25 % Nyco-denz in ice-cold sucrose medium) and the tubes centri-fuged at 106 000 g in a Kontron TFT-70 rotor for 90min. After centrifugation, the upper debris layer wasremoved using a spatula and the Nycodenz was re-moved by pipette. The remaining pure peroxisomepellet was dissolved in 1 mL ice-cold sucrose mediumand used for both freeze-etching and TEM studies.

For freeze-etching the peroxisomal pellets were dis-persed in 20 % glycerol, applied to gold specimencarriers, and frozen in propane precooled in liquidnitrogen. After freezing, the specimens were stored inliquid nitrogen until freeze-cleaved at –100 °C andetched for 2 min on a Balzers BAF 400 freeze-etchingapparatus. They were then unidirectionally shadowed(45°) with a 2 nm thick platinum layer and a 20 nmthick carbon layer (90°). The thickness of each layerwas automatically measured with a Balzers quartzcrystal Film Thickness Monitor QSG 301. The replicaswere cleaned in alternating baths of 10 % perchloricacid and distilled water, and picked up on grids forexamination in the electron microscope. The freeze-etching micrographs were used to quantitate the IMPson the opposing fracture faces, the concave face incontact with the protoplasm (PF) and the convex facein contact with the endoplasmic space (EF). The meansize of IMPs on both fracture faces was measured on aLeitz ASM 68K morphometrical analyzer and the num-ber of IMPs per square µm (µm2) was calculated.

For the TEM studies pure peroxisomal pellets werefixed in 1.5 % glutaraldehyde for 5 min, and subse-quently postfixed in 1 % OsO4 for 1 h, before dehydra-tion and embedding in Epon 812. Ultrathin sections

Biology of the Cell 92 (2000) 317–329 319


were contrasted with uranyl acetate and lead citrateand examined in the electron microscope.

2.4. Biochemical methods

2.4.1. RNA purificationSamples from the diaphragm, heart and liver were

frozen in liquid N2 immediately after dissection andstored at –80 °C. Total cellular RNA was purified usingthe acid guanidinium thiocyanate-phenol method(Chomczynski and Sacchi, 1987). RNA concentrationswere determined by measuring ultraviolet absorbanceat 260 nm.

2.4.2. Preparation of hybridization probesThe appropriate DNA fragments were excised from

plasmids by restriction enzymes, and the purified frag-ments were 32P-labeled using the oligo-labeling tech-nique (Feinberg and Vogelstain, 1984). The resultingspecific activities ranged from 1 to 5 × 109 cpm/mg.The probes were purified fragments of: CPT-I, 2600 bpEcoRI fragment of pBK2-CPT-I (Esser et al., 1993);partial CPT-II, 1600 bp XhoI/ XbaI fragment of PbKs-CPT-II.4 (Woeltje et al., 1990) and FAO, 1400 bp PstIfragment of pMJ125 (Miyazawa et al., 1987).

2.4.3. Hybridization and analysis of mRNASlot- and northern blotting of RNA onto nylon

membranes was carried out as previously described(Aasland et al., 1988; Frøyland et al., 1997b). Hybrid-ization to immobilized RNA was carried out in thepresence of 50 % formamide, 5 × SSC (0.75 M NaCl, 75mM trisodium citrate dihydrate), 200 mg/mL heat-denatured herring sperm DNA, 0.1 % SDS, 25 mMsodium phosphate (pH 6.4) and 8.25 % dextran sul-phate at 42 °C for 24–48 h. Filters were stringentlywashed twice in 0.2×SSC, 0.1 % SDS, 0.1 % Na4P2O7 at65 °C. When the filters were to be rehybridized, thebound probe was first stripped off by incubation of thefilters in 0.1 % SDS at 95–100 °C for 7 min. Kodak XAR5X-ray films were exposed to the membranes at –80 °Cin the presence of intensifying screens. Autoradio-grams were analyzed by densiometric scanning usingan LKB Ultrogel laser densiometer (Bromma, Sweden).The relative level of mRNA expression was estimatedas the amount of radioactive probe hybridized to eachsample of RNA relative to the level of 28S rRNA ineach sample.

2.4.4. Statistical analysisStudent’s t-test was applied to the areal fractions of

mitochondria and fat droplets in the individual experi-mental groups. Treatment means were compared byone-way analysis of variance. Fisher’s Least Squares

Difference test was used to determine the significanceof differences between the means at the 95 % confi-dence interval.

3. RESULTS

3.1. Skeletal muscle

PPDA-stained transverse sections of diaphragm andsoleus muscle from rats easily permits distinction be-tween type I and type II fibers. The mitochondria-rich,small-diameter fiber group is presumed to be type Ifibers (oxidative); and the mitochondria-poor, largediameter group is presumed to be type II fibers (glyco-lytic). In type I fibers the mitochondria formed chainsbetween the myofibrils and aggregations just beneaththe sarcolemma. The type II fibers contained moreevenly distributed, smaller mitochondria than the typeI fibers.

In both diaphragm and soleus muscle, type I fibersrevealed an increase in areal fraction occupied bymitochondria and polydispersion of mitochondria aftertreatment with 3-thia fatty acid and EPA, while in theDHA-treated group the mean areal fraction of mito-chondria and the distribution of mitochondrial profilearea was close to that of the controls (tables I, II) (figure1). The mean number of mitochondrial profiles in theconstant field of view measured in the different experi-mental groups was relatively constant within eachmuscle fiber type in both diaphragm and soleusmuscle. The distribution pattern of mitochondrial pro-files in type I fibers of the EPA and the 3-thia fatty acidtreated group were very similar (figure 1). In addition,the areal fractions of fat droplets in the type I fibers inthe EPA and 3-thia fatty acid treated groups werelower than in the DHA and control groups (tables I, II).Negative correlations (r2 = 0.96; P < 0.02 and r2 = 0.89;P < 0.05) were found between the areal fractions of fatdroplets and mitochondria in type I fibers from ratdiaphragm (table I) and soleus muscle, respectively(table II).

In the type II fibers the 3-thia fatty acid inducedgrowth of mitochondria and polydispersion of mito-chondria (tables I, II) (figure 1). In the EPA, DHA andcontrol groups, the mean areal fraction of mitochondriawas of the same magnitude (tables I, II). In addition, themean area of mitochondrial profiles and the polydis-persion of the mitochondria showed a closely similarpattern (figure 1).

CPT-I is generally accepted as being the rate-limitingenzyme of mitochondrial �-oxidation also in extrahe-patic tissue. Only DHA treatment significantly in-creased the gene expression of this enzyme indiaphragm compared to the control (table III). On theother hand, the mRNA levels of CPT-II were onlyinduced by 3-thia fatty acid treatment. No changes of



gene expression of CPT-I and CPT-II were observedwith EPA treatment (table III).

Although peroxisomes could not be visualized indiaphragm tissue, mRNA signals of the rate-limitingenzyme in peroxisomal �-oxidation, FAO, were de-tected (table III). The mRNA level of FAO was notaffected by EPA or DHA. However, the gene expres-sion of FAO was significantly increased after 3-thiafatty acid treatment.

In summary, in the skeletal muscle the morphologi-cal changes of the mitochondria did not correlate withthe gene expression of CPT-I and CPT-II. The resultsalso suggest that type I fibers are more liable toundergo morphological changes (i.e., in mitochondriaand lipid droplets) than type II fibers.

3.2. Heart

In order to study these phenomena in more detail inanother tissue with high aerobic metabolism, the inves-tigation was extended to heart tissue. Comparison ofheart sections from rats fed a normal pellet diet (con-trol group) and rats fed EPA and 3-thia fatty acidrevealed that the animals of the last two groups had agenerally normal structure and organization of theorganelles of the myocardial cells, except that themitochondria were distinctly enlarged (figure 2). ThemRNA levels of mitochondrial CPT-II decreased after3-thia fatty acid treatment (table III) (figure 3). Gene-expression of peroxisomal FAO was also detected inheart (table III), but the mRNA levels remained un-changed after EPA and DHA treatment and decreasedafter 3-thia fatty acid (table III).

3.3. Liver

Depending on the dose, 3-thia fatty acids differentlyaffected the volume fraction, mean number and size ofmitochondria in the liver of rats (Kryvi et al., 1990). Ingeneral, when the areal fraction of the mitochondriawas increased the number was only slightly affected.In contrast, when the areal fraction of mitochondriawas decreased the number of this organelle was sig-nificantly increased. Under both conditions, the arealfraction of fat droplets decreased. In contrast to DHA,EPA administration increased the areal fraction of mi-tochondria and lowered the areal fraction of fat drop-lets (Frøyland et al., 1997a). Development of mega-mitochondria and increased areal fraction ofmitochondria were observed in the hepatocytes afterEPA, but no DHA exposure (table IV) (figure 2). How-ever, administration of methyl-EPA induced a highergrowth of mitochondria than EPA (figure 2). In agree-ment with earlier studies (Madsen et al., 1999c), treat-ment with EPA, but not with DHA, increased themRNA levels of CPT-I (table III) (figure 3), while bothEPA and DHA increased the mRNA levels of CPT-II(table III). Administration of TTA for 1 week marginallyincreased the CPT-I mRNA (1.3-fold), while CPT-IImRNA increased more than three-fold (table III) (figure3).

In accordance with earlier findings (Frøyland et al.,1997a), morphometric measurements of randomly se-lected hepatocytes showed that DHA, but not EPAtreatments increased the areal fraction (%) of peroxi-somes (table V). The 3-thia fatty acid treatment resultedin an increase in both the areal fraction (11-fold; control

Table I. Diaphragm muscle.

Type I muscle fibers Type II muscle fibersFiber area (µm_) Areal fraction of

mitochondria(%)

Profile area ofmitochondria

(µm_)

Areal fraction offat droplets (%)

Fiber area (µm_) Areal fraction ofmitochondria

(%)

Areal fraction offat droplets (%)

3-thia fatty acid 1 379 ± 386 19.5 ± 3.4 0.23 ± 0.11 0.28 ± 0.18 3 129 ± 806 9.2 ± 1.7 0EPA 1 681 ± 734 13.2 ± 2.1 0.19 ± 0.14 1.75 ± 0.76 3 557 ± 907 5.3 ± 1.9 0DHA 1 243 ± 595 10.2 ± 2.3 0.12 ± 0.08 2.0 ± 0.34 3 082 ± 840 4.9 ± 2.7 0Control 1 395 ± 547 9.9 ± 1.3 0.12 ± 0.05 2.40 ± 0.68 3 497 ± 751 4.5 ± 2.2 0

Table II. Soleus muscle.

Type I muscle fibers Type II muscle fibersFiber area (µm_) Areal fraction of

mitochondriaProfile area ofmitochondria

Areal fraction offat droplets

Fiber area Areal fraction ofmitochondria

Areal fraction offat droplets

(%) (µm_) (%) (µm_) (%) (%)

3-thia fatty acid 2 480 ± 990 23.7 ± 4.6 0.22 ± 0.18 0.1 ± 0.05 3 309 ± 673 8.7 ± 3.1 0EPA 2 397 ± 1 063 14.5 ± 4.0 0.19 ± 0.14 2.5 ± 1.2 3 071 ± 570 4.9 ± 1.3 0DHA 2 510 ± 1 235 9.3 ± 3.1 0.13 ± 0.12 3.8 ± 1.1 325 ± 840 5.2 ± 2.1 0Control 2 362 ± 1 185 11.9 ± 2.9 0.11 ± 0.09 4.5 ± 1.3 3 265 ± 687 5.6 ± 2.7 0



1.4 versus 3-thia fatty acid 15.4) and the mean area ofperoxisomes (two-fold; control 10.22 versus 3-thia fattyacid 0.44). Confirming earlier studies (Madsen et al.,1999c), the gene expression of FAO, the rate-limitingenzyme of peroxisomal fatty acid oxidation, was sig-nificantly increased after EPA and DHA treatment(table III). The effect was marginal, however, in com-parison with 3-thia fatty acid treatment.

In order to further investigate the morphologicalchanges in the peroxisomes after 3-thia fatty acidtreatment, rats were treated with a strong peroxisomeproliferating 3-thia fatty acid. Transmission electronmicrographs (TEM) of purified peroxisomes from con-trol and 3-thia fatty acid treated rats showed that thecontamination of mitochondria was low (figure 4A). Inagreement with earlier findings (Kryvi et al., 1990), thefrequency of peroxisomes with dense cores decreasedas a result of feeding with 3-thia fatty acid (figure 4A).

Freeze-etching of peroxisomes was performed andelectron micrographs of the replicas were used tomeasure the distribution, frequency, shape and size ofthe intramembranous particles (IMP). The freeze-etched replicas show two kinds of fracture faces: aconcave protoplasmic face (figure 4C, E) and a convexendoplasmic face (figure 4D, F). Both the protoplasmicand endoplasmic aspects appeared as smooth facesthat were studded with randomly distributed IMPs.IMPs were more frequent on the protoplasmic facethan on the endoplasmic face in peroxisomes fromboth normal and 3-thia fatty acid-treated rats (table VI).In the control rats there were about ten times as manyIMPs on the protoplasmic face as on the endoplasmicface (table VI). The total number of IMPs per µm2

increased two-fold on the protoplasmic face after 3-thiafatty acid treatment (table VI). On the endoplasmic face,however, the number of IMPs was somewhat reduced

Figure 1. Size distribution histograms of mitochondrial areas of type I and type II muscle fiber types in the rat diaphragm (A, B) and soleusmuscle (C, D) based on morphometric measurements of mitochondria on 25 micrographs from each treatment group after treatment for12 weeks with 3-thia fatty acid (TTA), EPA, DHA. Control animals received CMC. Ordinate, classes of mitochondral area, each representing0.083 square micrometres (µm2); abscissa, percentage of mitochondria in each class.



after feeding with 3-thia fatty acid (table VI). The meanarea of IMPs was, however, not increased after 3-thiafatty acid treatment (table VI).

The freeze-etching studies revealed a heterogeneouspopulation of IMPs (figure 5). A number of the IMPswere apparently composed of two, and others of threesub-units, while others were elongated in shape. SomeIMPs showed a central indentation, while others pos-sessed a small central protrusion (data not shown). Themean IMP size on the protoplastic face of peroxisomeswas smaller in the hepatocytes of 3-thia fatty acid-treated rats than in the controls (figure 5A, B). 3-thiafatty acids thus seem to have an inductive effect on apopulation of small IMPs and also cause the largestIMPs to disappear (figure 5A, B). On the endoplasmicface there was a 32 % decrease in the mean IMP size onthe peroxisomes from 3-thia fatty acid-treated rats,compared to those of the controls (figure 5A, B). In thiscase most of the particles were absent.

4. DISCUSSION

This study has demonstrated that DHA had no effecton the mean size of mitochondria in neither type I nortype II fibers, whereas EPA and 3-thia fatty acidsinduced growth of mitochondria in type I fibers. The3-thia fatty acids also affect the mitochondrial biogen-esis in type II fibers. In liver and heart EPA, but notDHA, induced growth and proliferation of mitchon-dria. 2-methyl-EPA cannot be oxidized in mito-chondira, and in hepatocytes this fatty acid analogue

increased the mitochondrial area more than EPA. The3-thia fatty acids also increased the mean size ofmitochondria in the heart. Thus, the order of mitochon-dria proliferating capacity towards the various fattyacids is 3-thia fatty acids > 2-methyl-EPA > EPA >>DHA. Evidently, fatty acids that are non-�-oxidizable(3-thia fatty acids > 2-methyl-EPA) or poorly oxi-dazable (EPA) may initiate a signal for mega-mitochondria formation and subsequentlyproliferation. Most likely, these fatty acids are enteringthe mitochondria, as they are substates for CPT-I andCPT-II (Madsen et al., 1998, 1999c). Inside the mito-chondria EPA are poorly oxidized due to the doublebounds, whereas 2-methyl-EPA and 3-thia fatty acidsare blocked for oxidation as they are probably notsubstates for acyl-CoA dehydrogenase (Neat et al.,1980; Flatmark et al., 1983; Berge and Hvattum, 1994).In this context it is of interest that DHA is a poorsubstrate for CPT-I and CPT-II, and therefore DHAprobably do not enter the mitochondria (Madsen et al.,1998, 1999c).

The hepatic triglyceride levels were reduced by EPA,2-methyl-EPA and 3-thia fatty acids, but not DHA(Kryvi et al., 1990; Frøyland et al., 1997a). 3-thia fattyacids and EPA also reduced triglyceride levels in theheart (Hexeberg et al., 1995). In the present study wehave found that 3-thia fatty acids and EPA reduce thesize and number of fat droplets in type I muscle fibersof soleus and diaphragm muscle. Thus, morphologicalchanges of mitochondria, and possibly also their fattyacid oxidation capacity (Madsen et al., 1998), seems to

Table III. Gene expression of carnitine palmitoyltransferase-I (CPT-I), carnitine palmitoyltransferase-II (CPT-II) and fatty acyl-CoAoxidase (FAO) in liver, heart and diaphragm of rats fed 3-thia fatty acid (TTA), eicosapentaenoic acid (EPA) and docosahexaenoic acid(DHA) for seven days compared to control*.

Control 3-thia fatty acid EPA DHA

DiaphragmFAO 1.0 ± 0.3a 2.5 ± 0.3b 0.6 ± 0.2a 0.8 ± 0.2a

CPT-I 1.0 ± 0.1a 1.3 ± 0.3a,b 1.0 ± 0.1a 1.5 ± 0.2b

CPT-II 1.0 ± 0.3a 1.3 ± 0.1b 0.9 ± 0.2a 1.1 ± 0.2a

HeartFAO 1.0 ± 0.2a 0.4 ± 0.1b 0.8 ± 0.1a 0.9 ± 0.1a

CPT-I l NDCPT-II 1.0 ± 0.1a 0.4 ± 0.2b 0.7 ± 0.4a 0.6 ± 0.5a

LiverFAO 1.0 ± 0.1a 6.3 ± 1.1c 1.7 ± 0.3b 2.3 ± 0.5b

CPT-I 1.0 ± 0.1a 1.6 ± 0.3b 1.6 ± 0.3b 1.2 ± 0.1a

CPT-II 1.0 ± 0.1a 3.9 ± 0.6c 1.3 ± 0.1b 1.7 ± 0.2b

* Rats were treated for seven days with 3-thia fatty acid (TTA) at a dose of 300 mg/day/kg body weight, and EPA and DHA wereadministered at a dose of 1 500 mg/day/kg body weight or they were fed with CMC. The enzyme activities were measured inhomogenates prepared from fresh individual diaphragms, hearts and livers, and results are expressed as means±SD of four animalsin each group. The mRNA were extracted from frozen tissues, and data obtained from hybridization of northern blot membranes.The relative mRNA levels were determined by densitometric scanning of autoradiograms and the different mRNA levels werenormalized to 28S rRNA (fold increase). Means of CMC treatment are set to 1.0. Data are given as means ± SD of four different rats.Means in a row with a different superscript are significantly different (ANOVA, P < 0.05).



be inversely correlated to the triglyceride content andfat droplets.

It has been reported that the magnitude of thetriglyceride lowering effect towards various fatty acidsis 3-thia fatty acids > 2-methyl-EPA > EPA (Willumsenet al., 1998; Vaagenes et al., 1999; Berge et al., 1999).DHA, however, has no triglyceridemic effects in rats(Willumsen et al., 1993, 1996; Frøyland et al., 1997a;Madsen et al., 1999c; Berge et al., 1999). On the otherhand, DHA is a more potent peroxisome proliferatingfatty acid than EPA (Frøyland et al., 1997a). Thus, wehave suggested that the mitochondrion is the principaltarget for nutritional and pharmacological control of

triglyceride metabolism (Frøyland et al., 1997a). Evi-dently, in hepatocytes increased mitochondial�-oxidation with a concomitant decrease in triglyceridesynthesis and secretion may be an important mecha-nism underlying the triglyceride lowering effect of

Figure 2. Size distribution histograms of mitochondrial areasfrom myocytes (A) and hepatocytes (B) based on morphometricmeasurements of mitochondria on 25 micrographs from eachtreatment group after treatment for 12 weeks with EPA, DHA (A)3-thia fatty acid (TTA), or (B) 2-methyl-EPA, EPA or DHA. Controlanimals received CMC. Ordinate, classes of mitochondral area,each representing 0.33 square micrometres (µm2); abscissa,percentage of mitochondria in each class.

Figure 3. Effects of EPA, DHA and 3-thia fatty acid (TTA) on CPT-IImRNA levels in rat liver, heart and diaphragm. Rats were treatedfor 7 days with TTA at a dose of 150 mg/day/kg body weight, EPAor DHA at a dose of 1500 mg/day/kg body weight suspended in0.5 % CMC. Control animals received CMC only. Total mRNA waspurified from each individual liver, heart and diaphragm. Northernblotting and hybridization to immobilized RNA were performed asdescribed in Materials and methods. The figure shows represen-tative mRNA levels of CPT-II.

Table IV. Morphometric parameters of mitochondria in hepa-tocytes from rats fed various fatty acids*.

Treatment Control EPA DHA

Mean area (µm2) 0.28 0.37 0.28Areal fraction (%) 14.1 21.8 18.2

* Rats were treated for three months with 3-thia fatty acid(TTA) at a dose of 300 mg/day/kg body weight, or EPA orDHA at a dose of 1000 mg/day/kg body weight, suspended in0.5 % CMC. Control animals received CMC only.

Table V. Morphometric parameters of hepatic peroxisomes inhepatocytes from rats fed different fatty acids*.

Treatment Control EPA DHA

Mean area (µm2) 0.222 0.214 0.217Areal fraction (%) 1.4 1.2 1.9

* Rats were treated for three months with 3-thia fatty acid(TTA) at a dose of 300 mg/ day/ kg body weight, or EPA orDHA at a dose of 1000 mg/day/kg body weight, suspended in0.5 % CMC. Control animals received CMC only.



Figure 4. Morphological changes of liver peroxisomes after treatment with a 3-thia fatty acid (TD). A–B. Transmission electronmicrographs of purified peroxisomes from control (A) and TD treated (B) rats. Rats were fed a standard chow diet with or without 0.3 % TDfor 5 days. The livers were removed from decapitated rats and homogenized as described in Materials and methods. Bar, 600 nm. ×18 000. C–F. Freeze-etch micrograph of the protoplasmic face (C, E) and endoplasmic face (D, F) of representative peroxisome controls(C, D) and TD-treated (E, F) rats. Bar, 100 nm × 136 000.



EPA, 3-thia fatty acids and fibrates (Willumsen et al.,1993, 1996; Frøyland et al., 1997a; Madsen et al., 1999c;Berge et al., 1999). This is in agreement with inducedhepatic gene-expression of CPT-I in rats treated with3-thia fatty acids and EPA, but not DHA.

Although morphological changes in diaphragmwere observed, the mRNA levels measured in dia-phragm were not significantly changed by 3-thia fattyacids and EPA. DHA, however, induced gene-expression of CPT-I in this tissue. Assuming that in-duced gene expression of CPT-I runs in parallel withincreased �-oxidation in skeletal muscles, it may have aminimal contribution in the regulation of triglyseridecontent in blood.

The CPT-II mRNA levels were increased in liver ofEPA, DHA and especially in 3-thia fatty acid treatedanimals. In diaphragm the mRNA levels of CPT-IIwere enhanced 30 % by 3-thia fatty acids, whereas nosignificant changes were observed by EPA and DHAadministration. In heart, however, the mRNA levelswere reduced in 3-thia fatty acid treated animals. Thus,regulation of fatty acid metabolism in heart and skel-etal muscle may be different. Indeed, the mRNA levelsof FAO, the rate-limiting enzyme in peroxisomal�-oxidation, was increased in liver and diaphragm by3-thia fatty acids. In heart, however, the mRNA levelsof FAO were decreased. By EPA and DHA feeding,gene-expression of FAO was induced in liver, but notin diaphragm and heart.

In previous studies it has been shown that theincreased activity of FAO, CPT-I and CPT-II protein inthe liver was caused by increased level of their respec-tive mRNAs, both in response to 3-thia fatty acids andEPA administration. Decreased degradation of FAOhas also been proposed to contribute to the increasedprotein levels (Horie and Suga, 1989). Moreover, it hasbeen reported that both clofibrate and partially hydro-genated fish oil in the liver mainly induces transcrip-tion of the FAO gene, whereas in the heart thetranslation process seems to be of importance in theregulation of the FAO activity (Kvannes et al., 1994,1995). This dual regulatory mechanism may well bemore pronounced in the heart than in diaphragm and

liver, as the steady-state mRNA levels of FAO andCPT-II was reduced in heart after administration of3-thia fatty acid. As the �-oxidation capacities arereported to increase (Hexeberg et al., 1995), othermechanisms than regulation at the level of mRNA maybe involved, most likely stabilization of FAO andCPT-II proteins.

The induction of hepatic peroxisomal proliferationand gene-expression of FAO by xenobiotics in the liver,appear to depend on the peroxisome-proliferating acti-vated receptor (PPARa) (Ren et al., 1996; Gonzales,1997; Peters et al., 1997; Aoyama et al., 1998). Studiesusing PPARa knock-out-mice have indeed demon-strated that the hepatic induction of FAO by fibratesand fish oil is dependent on PPARa activation (Ren etal., 1997; Aoyoma et al., 1998).

In skeletal muscle, however, PPARa is expressed atlow levels (Schoonjans et al., 1996). PPARa mRNA ispredominantly expressed in tissues with high fattyacid catabolism, such as the liver and heart(Schoonjans et al., 1996). In contrast to the liver, heartmRNA levels were unchanged by EPA and DHA ad-ministration and actually reduced by 3-thia fatty acid.Although PPARa has been shown to be activated byhypolipidemic drugs, such as 3-thia fatty acids andpolyunsaturated fatty acids in hepatocytes (Forman etal., 1997; Raspé et al., 1999), activation of PPARa inheart and skeletal muscle by 3-thia fatty acids, EPA andDHA, is thus unlikely to take place.

In summary, in the liver, EPA and the 3-thia fattyacids share numerous important features: they all havea hypolipidemic effect, they induce growth and prolif-eration of mitochondria concomitant with the induc-tion of fatty acid oxidation and PPARa activated genes.It is therefore likely that PPARa activation by 3-thiafatty acids, EPA and DHA induce gene expression ofboth mitochondrial and peroxisomal genes as well asperoxisomal proliferation. However, DHA does notinduce growth of mitochondria. Thus, activation of thisreceptor might also be necessary, but not sufficient, toinduce growth of mitochondria. As to the effects of the3-thia acid TD on the liver peroxisomes, we haveshown that significant changes take place both in

Table VI. Morphometric values of hepatic intramembrane particles (IMPs) in hepatic peroxisomal membranes from control and 3-thiafatty acid (TD) treated rats*.

Fracture face Endoplasmic face Protoplasmic faceTreatment Control 3-thia fatty acid Control 3-thia fatty acid

Mean area of IMPs (nm2) 133 ± 50 91 ± 29 70 ± 24 58 ± 20Mean number of IMPs per µm2 of peroxisome membrane 232 ± 115 182 ± 140 2586 ± 703 5620 ± 1735

* Rats were fed a standard chow diet with or without 0.3 % TD. The livers were removed from decapitated rats and homogenized asdescribed in Materials and methods. Freeze-etching was performed and electron micrographs of the protoplasmic face and theendoplasmic face of a representative peroxisome control and a TD treated rat were used to measure the distribution, the mean areaof IMPs and the mean number of IMPs per µm2.



peroxisomal size and in the number and distribution ofmembrane-associated particles.

In conclusion, the strongly peroxisome-proliferating3-thia fatty acid, EPA and DHA affect the gene-expression and activity of mitochondrial and peroxiso-mal enzymes in different ways in diaphragm muscle,heart and liver. Our study indicates that EPA and DHAhave different effects on mitochondrial biogenesis inskeletal muscle, heart and liver.

Acknowledgments. This work was supported by theResearch Council of Norway. The authors are obligedto Nina Ellingsen, Teresa Cieplinska, Svein Krüger andKari Williams for technical assistance, Dr. Skattebøl,Norsk Hydro AS for providing DHA, EPA and2-methyl EPA, Dr. T. Hashimoto for the fatty acyl-CoAoxidase cDNA and Dr. V. Esser and Dr. J.D. McGarryfor the carninitine palmitoyltrasferase cDNAs.

REFERENCES

Aasland, R., Lillehaug, J.R., Male, R., Jøsendal, O., Varhaug,, J.E.,Kleppe, K., 1988. Expression of oncogenes in thyroid tumours:coexpression of c-erB2/neu and c-erbB. Br. J. Cancer 57, 358–364.

Aarsland, A., Berge, R.K., Bremer, J., Aarsaether, N., 1990. The hypolipi-demic peroxisome proliferating drug bis(carboxymethylthio)-1,10

decane, a dicarboxylic metabolite of tadenol, is activated to anacylcoenzyme A thioester. Biochim. Biophys. Acta 1033, 176–183.

Aoyama, T., Peters, J.M., Irtani, N., Nakajima, T., Furihata, K., Hash-imoto, T., Gonzalez, F.J., 1998. Altered constitutive expression offatty acid metabolizing enzymes in mice lacking the peroxisomeproliferator-activated receptor a (PPARa). J. Biol. Chem. 273,5678–5684.

Asiedu, D.K., Skorve, J., Willumsen, N., Demoz, A., Berge, R.K., 1993.Early effects of mitochondrial and peroxisomal b-oxidation by thehypolipidemic 3-thia fatty acid in rat livers. Biochim. Biophys. Acta1166, 73–76.

Asiedu, D.K., Demoz, A., Skorve, J., Grav, H.J., Berge, R.K., 1995. Acutemodulation of rat hepatic metabolism by sulfur-substituted fattyacid analogues. Biochem. Pharmacol. 49, 1013–1022.

Asiedu, D.K., Frøyland, L., Vaagenes, H., Lie, Ø, Demoz, A., Berge, R.K.,1996. Long-term effect of tetradecylthioacetic acid: A study on lipidprofile and fatty acid composition and oxidation in different ratorgans. Biochim. Biophys. Acta 1300, 86–96.

Berge, R.K., Hvattum, E., 1994. Impact of cytochrome P450 system onlipoprotein metabolism. Effect of abnormal fatty acids (3-thia fattyacids). Pharmacol. Ther. 61, 345–383.

Berge, R.K., Aarsland, A., Kryvi, H., Bremer, J., Aarsaether, N., 1989.Alkylthioacetic acids (3-thia fatty acids) - a new group of non-b-oxidizable, peroxisome inducing fatty acid analogues. I. A study onthe structural requirements for proliferation of peroxisomes andmitochondria in rat liver. Biochim. Biophys. Acta 1004, 345–356.

Berge, R.K., Madsen, L., Vaagenes, H., Tronstad, K.J., Gottlicher, M.,Rustan, A.C., 1999. In contrast to docosahexaenoic acid, eicosapen-taenoic acid and hypolipidemic derivatives reduce hepatic synthesisand secretion of triacylglycerol by decreased diacylglycerol acyl-

Figure 5. Size distribution histo-gram of integrated membrane par-ticles (IMP) in peroxisomalmembranes based on morphometricmeasurements of peroxisomes fromthe protoplasmic face (A, B) and theendoplasmic face (C, D) of a repre-sentative peroxisome control (A, C)and 3-thia fatty acid (TD) treated (B,D) rats. Rats were fed a standardchow diet with or without 0.3 % TDfor 5 days. The livers were removedfrom decapitated rats and homog-enized as described in Materials andmethods. Ordinate, IMP area insquare micrometres; abscissa, totalnumber of IMP.



transferase activity and stimulation of fatty acid oxidation. Biochem.J. 343, 191–197.

Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isolationby acid guanidinium thiocyanate-phenol-chloroform extraction.Anal. Biochem. 162, 156–159.

De Duve, C., Pressman, B.C., Gianetto, R., Wattiaux, R., Applemans, F.,1955. Intracellular distribution patterns of enzymes in rat livertissue. Biochem. J. 60, 604–617.

Esser, V., Britton, C.H., Weis, B.C., Foster, D.W., McGarry, J.D., 1993.Cloning, sequencing., expression of a cDNA encoding rat livercarnitine palmitoyltransferase I. J. Biol. Chem. 268, 5817–5822.

Feinberg, A.P., Vogelstain, B., 1984. A technique for radiolabeling DNArestriction endonuclease fragments to high specific activity. Anal.Biochem. 137, 266–267.

Flatmark, T., Christiansen, E.N., Kryvi, H., 1983. Evidence for a negativemodulating effect of erucic acid on the peroxisomal beta-oxidationenzyme system and biogenesis in rat liver. Biochim. Biophys. Acta753, 460–466.

Forman, B.M., Chen, J., Evans, R.M., 1997. Hypolipidemic drugs,polyunsaturated fatty acids and eicosanoids are ligands for peroxi-some proliferator-activated receptors a and δ. Proc. Natl. Acad. Sci.USA 94, 4312–4317.

Frøyland, L., Asiedu, D.K., Vaagenes, H., Garras, A., Lie, Ø, Totland,G.K., Berge, R.K., 1995. Tetradecylthioacetic acid which was incorpo-rated into very low density lipoprotein, changed the fatty acidcomposition and reduced plasma lipids in cholesterol-fed hamsters.J. Lipid Res. 361, 2529–2540.

Frøyland, L., Vaagenes, H., Asiedu, D.K., Garras, A., Lie, Ø, Berge, R.K.,1996. Chronic administration of eicosapentaenoic acid and docosa-hexaenoic acid as ethyl esters reduced plasma cholesterol andchanged the fatty acid composition in rat blood and organs. Lipids31, 169–178.

Frøyland, L., Madsen, L., Vaagenes, H., Totland, G.K., Auwerx, J.,Kryvi, H., Staels, B., Berge, R.K., 1997a. Mitochondrion is theprincipal target for nutritional and pharmacological control oftriglyceride metabolism. J. Lipid Res. 38, 1851–1858.

Frøyland, L., Madsen, L., Sjursen, W., Garras, A., Lie, Ø, Songstad, J.,Rustan, A.C., Berge, R.K., 1997b. Effect of 3-thia fatty acids on thelipid composition of rat liver, lipoproteins and heart. J. Lipid Res. 38,1522–1534.

Garras, A., Asiedu, A.K., Berge, R.K., 1995. Subcellular localisation andinduction of NADH-sensitive acetyl-CoA hydrolase and propionyl-CoA hydrolase activities in rat liver under lipogenic conditions aftertreatment with sulfur-substituted fatty acids. Biochim. Biophys. Acta1255, 154–160.

Grønn, M., Christiansen, E.N., Hagve, T.A., Christophersen, B.O., 1992.Effects of dietary purified eicosapentaenoid acid (20:5 (n-3)) anddocisahexaenoic acid (22:6 (n-3)) on fatty acid desaturation andoxidation in isolated rat liver cells. Biochim. Biophys. Acta 1123,35–43.

Gonzalez, F.J., 1997. Recent update on the PPAR a-null mouse. Bio-chimie 79, 130–144.

Gundersen, H.J.G., Bendtsen, T.F., Korbo, L., Marcussen, N., Møller, A.,Nielsen, K., Nyengaard, J.R., Pakkenberg, B., Sørensen, F.B., Vest-erby, A., West, M.J., 1988. Some new, simple and efficient stereologi-cal methods and their use in pathological research and diagnosis.APMIS 96, 379–394.

Harris, W.S., 1989. Fish oils and plasma lipid and lipoprotein metabo-lism in humans: A critical review. J. Lipid Res. 30, 785–807.

Hexeberg, S., Frøyland, L., Asiedu, D.K., Demoz, A.,, Berge, R.K., 1995.Tetradecylthioacetic acid reduces the amount of lipid droplets,induces megamitochondria formation and increases the fatty acidoxidation in rat heart. J. Mol. Cell Cardiol. 27, 1851–1857.

Horie, S., Suga, T., 1989. Turnover of fatty accyl-CoA oxidase in the liverof rats fed on a partially hydrogenated marine oil. Lipids 24,890–894.

Ikeda, I., Cha, J.Y., Yanagita, T., Nakatani, N., Oogami, K., Imaizumi, K.,Yazawa, K., 1998. Effects of dietary alpha linolenic, eicosapentaenoic

and docosahexaenoic acids on hepatic lipogenesis and beta-oxidation in rats. Biosci. Biotechnol. Biochem. 62, 675–680.

Korneliussen, H., 1972. Identification of muscle fiber types in ‘semithin’sections stained with p-phenylene-diamine. Histochemie 32, 95–98.

Kryvi, H., Aarsland, A., Berge, R.K., 1990. Morphologic effects ofsulphur-substibuted fatty acids on rat hepatocytes with specialreference to proliferation of peroxisomes and mitochondria. J. Struct.Biol. 103, 257–265.

Kvannes, J., Eikhom, T.S., Flatmark, T., 1994. The peroxisomal beta-oxidation enzyme system of rat heart. Basal level and effects of theperoxisome proliferator clofibrate. Biochim. Biophys. Acta 1201,203–216.

Kvannes, J., Eikhom, T.S., Flatmark, T., 1995. On the mechanism ofstimulation of peroxisomal �-oxidation in rat heart by partiallyhydrogenated fish oil. Biochim. Biophys. Acta 1255, 39–49.

Madsen, L., Frøyland, L., Dyrøy, E., Helland, K., Berge, R.K., 1998.Docosahexaenoic and eicosapentaenoic acid are differently metabo-lized in rat liver during mitochondria- and peroxisome proliferation.J. Lipid Res. 39, 583–594.

Madsen, L., Berge, R.K., 1999a. 3-thia fatty acid treatment, in contrast toeicosapentaenoic acid and starvation, induces gene expression ofcarnitine palmitoyltransferase-II in rat liver. Lipids 34, 447–456.

Madsen, L., Garras, A., Asins, G., Serra, D., Hegardt, F.G., Berge, R.K.,1999b. Mitochondrial 3-hydroxy-3-methylglutaryl Coenzyme A syn-thase and carnitine palmitoyltransferase II as potential control sitesfor ketogenesis during mitochondrion and peroxisome proliferation.Biochem. Pharmacol. 57, 1011–1019.

Madsen, L., Rustan, A.C., Vaagenes, H., Berge, K., Dyrøy, E., Berge,R.K., 1999c. Eicosapentaenoic and docosahexaenoic acids affectmitochondrial and peroxisomal fatty acid oxidation in relation tosubstrate preference. Lipids 34, 951–963.

McGarry, J.D., Brown, N.F., 1997. The mitochondrial carnitine palmi-toyltransferase system. From concept to molecular analysis. Eur. J.Biochem 244, 1–14.

Miyazawa, S.H., Hayashhi, M., Hijikata, N., Ishii, S., Furuta, H.,Kagamiyama, T., Osumi, T., Hashimoto, T., 1987. Complete nucle-otide sequence of cDNA and predicted amino acid sequence of ratacyl-CoA oxidase. J. Biol. Chem. 262, 8131–8137.

Neat, C.E., Thomassen, M.S., Osmundsen, H., 1980. Induction ofperoxisomal beta-oxidation in rat liver by high-fat diets. Biochem. J.186, 369–371.

Peters, J.M., Hennuyer, N., Staels, B., Fruchart, J.C., Fivet, C., Gonzalez,F.J., Auwerx, J., 1997. Alterations in lipoprotein metabolism inperoxisome proliferator-activated receptor a-deficient mice. J. Biol.Chem. 272, 27307–27312.

Ramsay, R.R., 1999. The role of the carnitine system in peroxisomal fattyacid oxidation. Am. J. Med. Sci. 318, 28–35.

Raspé, E., Madsen, L., Lefebvre, A.M., Leitersdorf, I., Gelman, L.,Peinado-Onsurbe, J., Dallongeville, J., Fruchart, J.C., Berge, R.,Staels, B., 1999. Modulation of rat liver apolipoprotein gene expres-sion and serum lipid levels by tetradecylthioacetic acid (TTA) viaPPARa activation. J. Lipid Res. 40, 2099–2110.

Ren, B., Thelen, A.P., Jump, D.B., 1996. Peroxisome proliferator-activated receptor a inhibits hepatic S14 gene transcription. Evidenceagainst the peroxisome proliferator-activated receptor a as themediator of polyunsaturated fatty acid regulation of s14gene tran-scription. J. Biol. Chem. 271, 17167–17173.

Ren, B., Thelen, A.P., Peters, J.M., Gonzalez, F.J., JumpD.B, 1997.Polyunsaturated fatty acid suppression of hepatic fatty acid syn-thase and S14 gene expression does not require peroxisomeproliferator-activated receptor a. J. Biol. Chem. 272, 26827–267832.

Schoonjans, K., Staels, B., Auwerx, J., 1996. The peroxisome proliferatoractivated receptors (PPARb) and their effects on lipid metabolismand adipocyte differentiation. Biochim. Biophys. Acta 1302, 93–109.

Skrede, S., Sørensen, H.N., Larsen, L.N., Steineger, H.H., Hovik, K.,Spydevold, O.S., Horn, R., Bremer, J., 1997. Thia fatty acids, metabo-lism and metabolic effects. Biochim. Biophys. Acta 1344, 115–131.



Surette, M.E., Whelan, J., Broughton, K.S., Kinsella, J.E., 1992. Evidencefor mechanisms of the hypotriglyceridemic effect of n-3 polyunsatu-rated fatty acids. Biochim. Biophys. Acta 1126, 100–205.

Spydevold, Ø, Bremer, J., 1989. Induction of peroxisomal b-oxidation in7800 C1 Morris hepatoma cells in steady state by fatty acid ana-logues. Biochim. Biophys. Acta 1003, 72–79.

Vaagenes, H., Madsen, L., Asiedu, D.K., Lillehaug, J.R., Berge, R.K.,1998. Early modulation of genes encoding peroxisomal and mito-chondrial b-oxidation enzymes by 3-thia fatty acids. Biochem.Pharmacol. 56, 1571–1582.

Vaagenes, H., Madsen, L., Dyrøy, E., Elholm, M., Stray-Pedersen, A.,Frøyland, L., Lie, Ø, Berge, R.K., 1999. Methylated eicosapentaenoicacid and tetradecylthioacetic acid: effects on fatty acid metabolism.Biochem. Pharmacol 58, 1133–1143.

Willumsen, N., Hexeberg, S., Skorve, J., Lundquist, M., Berge, R.K.,1993. Docosahexaenoic acid shows no triglyceride lowering effect

but increases the peroxisomal fatty acid oxidation in liver of rats. J.Lipid Res. 34, 13–22.

Willumsen, N., Vaagenes, H., Asiedu, D., Lie, Ø, Rustan, A.C., Berge,R.K., 1996. Eicosapentaenoic acid but not docosahexaenoic acid(both as ethyl esters) increases mitochondrial fatty acid oxidationand upregulates 2,4-dienoyl-CoA reductase gene expression. Apotential mechanism for the hypolipidemic action of fish oil in rats.Lipids 31, 579–592.

Willumsen, N., Vaagenes, H., Holmsen, H., Berge, R.K., 1998. On theeffect of 2-deuterium- and 2-methyl-eicosapentaenoic acid deriva-tives on triglycerides, peroxisomal beta-oxidation and platelet ag-gregation in rats. Biochim. Biophys. Acta 1369, 193–203.

Woeltje, K.F., Esser, V., Weis, B.C., Cox,, W.F., Schroeder, J.G., Liao, S.-T.,Foster, D.W., McGarry, J.D., 1990. Inter-tissue and inter-speciescharacteristics of the mitochondrial carnitine palmitoyltransferaseenzyme system. J. Biol. Chem. 265, 10714–10719.



Proliferation of mitochondria and gene expression of carnitine palmitoyltransferase and fatty...

Documents

Transcript of Proliferation of mitochondria and gene expression of carnitine palmitoyltransferase and fatty...